Field of Invention
The present invention pertains to the computing sciences generally, and, more specifically to an apparatus and method for vector multiplication with accumulation in large register space.
Background
FIG. 1 shows a high level diagram of a processing core 100 implemented with logic circuitry on a semiconductor chip. The processing core includes a pipeline 101. The pipeline consists of multiple stages each designed to perform a specific step in the multi-step process needed to fully execute a program code instruction. These typically include at least: 1) instruction fetch and decode; 2) data fetch; 3) execution; 4) write-back. The execution stage performs a specific operation identified by an instruction that was fetched and decoded in prior stage(s) (e.g., in step 1) above) upon data identified by the same instruction and fetched in another prior stage (e.g., step 2) above). The data that is operated upon is typically fetched from (general purpose) register storage space 102. New data that is created at the completion of the operation is also typically “written back” to register storage space (e.g., at stage 4) above).
The logic circuitry associated with the execution stage is typically composed of multiple “execution units” or “functional units” 103_1 to 103_N that are each designed to perform its own unique subset of operations (e.g., a first functional unit performs integer math operations, a second functional unit performs floating point instructions, a third functional unit performs load/store operations from/to cache/memory, etc.). The collection of all operations performed by all the functional units corresponds to the “instruction set” supported by the processing core 100.
Two types of processor architectures are widely recognized in the field of computer science: “scalar” and “vector”. A scalar processor is designed to execute instructions that perform operations on a single set of data, whereas, a vector processor is designed to execute instructions that perform operations on multiple sets of data. FIGS. 2A and 2B present a comparative example that demonstrates the basic difference between a scalar processor and a vector processor.
FIG. 2A shows an example of a scalar AND instruction in which a single operand set, A and B, are ANDed together to produce a singular (or “scalar”) result C (i.e., AB=C). By contrast, FIG. 2B shows an example of a vector AND instruction in which two operand sets, A/B and D/E, are respectively ANDed together in parallel to simultaneously produce a vector result C, F (i.e., A.AND.B=C and D.AND.E=F). As a matter of terminology, a “vector” is a data element having multiple “elements”. For example, a vector V=Q, R, S, T, U has five different elements: Q, R, S, T and U. The “size” of the exemplary vector V is five (because it has five elements).
FIG. 1 also shows the presence of vector register space 104 that is different that general purpose register space 102. Specifically, general purpose register space 102 is nominally used to store scalar values. As such, when, the any of execution units perform scalar operations they nominally use operands called from (and write results back to) general purpose register storage space 102. By contrast, when any of the execution units perform vector operations they nominally use operands called from (and write results back to) vector register space 107. Different regions of memory may likewise be allocated for the storage of scalar values and vector values.
Note also the presence of masking logic 104_1 to 104_N and 105_1 to 105_N at the respective inputs to and outputs from the functional units 103_1 to 103_N. In various implementations, only one of these layers is actually implemented—although that is not a strict requirement. For any instruction that employs masking, input masking logic 104_1 to 104_N and/or output masking logic 105_1 to 105_N may be used to control which elements are effectively operated on for the vector instruction. Here, a mask vector is read from a mask register space 106 (e.g., along with input data vectors read from vector register storage space 107) and is presented to at least one of the masking logic 104, 105 layers.
Over the course of executing vector program code each vector instruction need not require a full data word. For example, the input vectors for some instructions may only be 8 elements, the input vectors for other instructions may be 16 elements, the input vectors for other instructions may be 32 elements, etc. Masking layers 104/105 are therefore used to identify a set of elements of a full vector data word that apply for a particular instruction so as to effect different vector sizes across instructions. Typically, for each vector instruction, a specific mask pattern kept in mask register space 106 is called out by the instruction, fetched from mask register space and provided to either or both of the mask layers 104/105 to “enable” the correct set of elements for the particular vector operation.
FIG. 3 shows a standard “schoolbook” multiplication process within a base 10 system. As observed in FIG. 3, each digit in a multiplicand 301 is multiplied by each digit in a multiplier 302 to create an array of partial products 303. Each partial product is aligned with the location of its respective multiplier digit. The aligned partial product terms are added together to produce multiplication result 304.
Note the presence of the carry terms 305. Carry terms 305_1 through 305_5 can be created not only when the partial product terms are added to produce the final result, but also, as part of the determination of each partial product term itself. For example, carry term 305_1 is created during the summation of the partial products, while, each of carry terms 305_2 through 305_4 is generated in determining a particular partial product.
In order to perform multiplication operations a processing core embedded on a semiconductor chip essentially performs mathematical operations that are similar to the multiplication processes discussed above. Specifically, partial product terms are generated, and, the partial product terms are added to produce a final result. In the case of vector instructions, however, carry terms can present problems.
For example, any “special logic circuitry” needed to recognize and account for any generated carry terms can become substantial in size as such logic circuitry would be needed for every element of the maximum vector size supported by the processor. Non-vector “integer” execution logic of a processor may be designed to use special “flags” and corresponding flag circuitry to handle carry terms. However, as integer operations are essentially scalar operations, only one instance of such circuitry needs to be implemented.
As such, a common processor design point for a processor that supports both integer and vector instructions is to design special flag circuitry for the integer instructions but not the vector instructions (or at least a limited version of flag circuitry for the vector instructions). Without the flag circuitry and its corresponding support for carry terms, the designers of a processor's vector instruction execution logic face the challenge of accounting for carry terms in their vector multiplication instruction execution logic by some other technique.